Visual perception is selective, and visual attention is the mechanism by which salient or high-priority objects are prioritized for awareness and action. Recent work has shown that attentional selection often operates on perceptual objects and not simply on spatial locations. In a multi-object scene involving partial occlusion, image regions must be grouped into coherent object representations prior to attentional deployment. The visual system employs a set of heuristics that serve to constrain object recognition based on principles of perceptual organization. Despite their importance to human navigation and behavior, knowledge of these heuristics and the neural mechanisms that underlie this process remain poorly understood. The proposed project will examine the grouping principles that are critical to object-based attentional selection, and will enumerate the dominance relations among these principles when they are consistent with competing groupings within a scene. The project also explores the neural implementation of these processes to extend our knowledge of the constraints built into the object recognition and visual attention systems. The combination of behavioral and neuroimaging methods outlined in this application will result in a better understanding of the neural circuitry that is responsible for efficient, goal-directed object recognition.